FR2853483A1 - Multimedia data e.g. video, packet transmission managing process, involves transmitting interruption message to client device, when size of data packet received from client device is greater than available memory space in server - Google Patents

Abstract

The process involves comparing a size of a data packet received from a client device with a size of available memory space in a server for storing the data packets. An interruption message to temporarily interrupt the transmissions of the data packets to the server is transmitted to the client device, when the size of the received data packet is greater than the available memory space in the server. Independent claims are also included for the following: (a) a communication apparatus including the device of management of multimedia data packets transmission between two communication apparatus (b) an information storage unit readable by a computer or a microprocessor comprising the instructions of computer program for executing the steps of the process of management of the transmission of multimedia data packets (c) a computer program loaded in a programmable apparatus and comprising a sequence of instructions or the logical code for executing the process of management of transmission of multimedia data.

Description

i

  The present invention relates to a method and a device for managing transmissions of multimedia data packets between at least two communication devices connected to one another.

  The technological advancement of digital communication devices such as television sets, pocket PCs, conventional PCs or any other miniaturized device is such that, in the coming years, it will most likely be possible to exchange data. multimedia type and, for example, audio and / or video between these different devices.

  Market research shows, moreover, that users are ready to buy such communication devices provided that the price is not too excessive and, above all, that the services provided are of good quality.

  Following the identification of this new market, numerous studies have been conducted on the miniaturization of electronic components, in order to obtain devices having high memory and computing capacity.

  In addition, many standardization bodies have established standards to unify the exchange of data between these devices with different characteristics. These standards, such as those of the IEEE 802 family, allow both the implementation and management of data transfer via a particular network.

  The IEEE 802.11 standard and its European counterpart HYPERLAN are dedicated to wireless LANs. They are particularly studied now because they are broadband networks for transmitting audio and / or video data between two machines that are geographically close.

  These standards are therefore particularly suitable for developing wireless local area networks within homes.

  The technical solutions currently studied in the context of the aforementioned standards applied to the local home networks make it possible to transmit digital data either between a so-called server machine and a so-called client machine (point-to-point transmission), or between the server and a group of clients, again between the server and several clients (multi-point communication) without any wires.

  In this environment, the data is stored on the server that centralizes all communications with the client (s). This server can also serve as a gateway to the outside world (Internet, television, camcorder ...).

  It should be noted that home LANs may be of a heterogeneous nature, i.e. they may, for example, consist partly of wireless links and partly of wired links.

  While many projects are studying the implementation of interactive television services and the exchange of information on the lnternet, few of them deal with problems related to the implementation of local home networks, namely: the implementation of solutions to obtain a quality of services acceptable to users.

  The transmission of data, for example, of video type between a server and a client machine through a local communication network, so as to allow the client machine to decode and visualize this video data is not an easy service to implement.

  Indeed, several problems arise: - the video that has been requested by the client machine via a prior request must be adapted to the context in which the transmission takes place, namely, for example, the bandwidth limited communication network, the possibility of loss of data packets and that bits are erroneous during this transmission, reduced processing and / or storage capacity of the client machine ...; it should also be taken into account that several client machines may simultaneously request the same video or even different videos; for this reason, using a so-called "best effort" strategy (known in English terminology as the "best effort") in which a packet of data is sent to each client machine in turn, some of these machines will receive a large number of data to be processed (decoding and visualization), while other machines will not receive enough to ensure a correct visualization of the decoded video: then we will speak respectively of advance client machines and late client machines; it should also be noted that the client machine is generally a computer or, at least, a machine having a computing capacity and a memory capacity, the memory in question being necessary for storing the reception, decoding and viewing the video; furthermore, it is necessary to allocate a portion of this memory to the temporary storage of the data packets received from the server pending their processing; however, when dealing with portable type client machines, such as handheld computers, personal digital assistants, and the like, the memory capacity of the machine is limited and the memory space dedicated to receiving data packets is generally smaller than the size of the video to be received, which poses the problem of exceeding the memory capacity of the machine.

  In view of the foregoing, it would therefore be advantageous to be able to simultaneously transmit a plurality of multimedia digital data entities, for example videos, to various other communication devices and to be able to manage the memory overflow 20 which may occur on the latter when receiving data.

  Document US Pat. No. 5,822,524 discloses a method for solving the problem of management of exceeding the memory capacity in a client machine.

  In this document, it is intended to transmit video data using a scheduling algorithm associated with each of the client machines by considering, at the server level, three states: - the NORMAL state in which the data packets are sent by the server at a determined rate, - the RUSH state in which the data packets are sent by the server at an accelerated rate, and - the PAUSE qualified state that no data packet is sent by the server. server.

  The change in the operating mode of this scheduling algorithm is piloted by the client according to the state of filling of the memory of the client machine allocated to the reception of data packets with respect to two thresholds, one noted H, beyond which the memory is considered as full and the other noted L, below which the memory is considered empty.

  The server is in the NORMAL state when the filling rate of the memory of the client machine is between the two aforementioned thresholds, this NORMAL state meaning for the client machine that the received data packets can be stored in the memory and that stored packets can be processed, i.e., decoded and viewed.

  When the filling rate of the memory is below the aforementioned threshold H, then a PAUSE message is sent to the server in order to interrupt the transmission of the packets to the client machine in question.

  When the filling rate of the memory falls below the aforementioned threshold H, then the client machine sends a RESTART message to the server so that it resumes the transmission of data packets to the latter.

  When the fill rate falls below the aforementioned threshold L, then a RUSH message is sent to the server, indicating to the server the need to quickly transmit data packets.

  The major disadvantage of this technique lies in the definition of the two thresholds H and L. Indeed, when it is desired to prevent the memory is empty, which could lead to a blocking of the decoder and therefore the algorithm of 25 display (the image might then be frozen on the display), it is necessary that the threshold L is high enough so that the processing time of the data packets remaining in the memory is greater than that required for sending the RUSH message to the server, time to add the time needed to receive a data packet.

  It should also be noted that this threshold L should not be too low because, indeed, the bandwidth of the communication network would be wasted by the frequent sending of messages RUSH, NORMAL and PAUSE by the client machine.

  We thus realize that the choice of threshold L is not particularly easy.

  The definition of the threshold H also poses a problem insofar as, as soon as the filling rate of the memory is greater than the threshold H, then the client machine sends the message PAUSE to the server but still continues to take into account data packets arriving.

  Since the client machine does not have prior knowledge of the size of the packets that it will receive, it is a priori impossible to define the threshold H by avoiding losing data packets that would arrive after the sending of the packet. PAUSE message.

  It can thus be seen that this type of method using two thresholds for managing the problem of exceeding the memory capacity of the client machine 15 requires transmitting data packets of constant size or, in any case, data packets whose size does not exceed a certain limit.

  The present invention aims to remedy at least one of the aforementioned drawbacks by proposing a method for managing transmissions of multimedia data packets between at least two communication devices connected to each other, one of said first apparatus receiving packets of data from at least one other apparatus, said second apparatus, characterized in that the method comprises the following steps performed by the first apparatus: - comparing the size of the last data packet received from the second apparatus with the size of the space memory available in the first device for its storage, - depending on the result of the comparison, decision on the transmission, in the direction of the second device, a message called interruption (PAUSE) to interrupt temporarily the transmission of data packets to the first device.

  Correlatively, the invention relates to a device for managing transmissions of multimedia data packets between at least two communication devices connected to one another, said first device receiving data packets from at least one other device, said second apparatus, characterized in that the first apparatus comprises: means for comparing the size of the last data packet received from the second apparatus with the size of the memory space available in the first apparatus for its storage; transmitting, in the direction of the second apparatus, a so-called interrupt message (PAUSE) for temporarily interrupting the transmission of data packets to the first apparatus, depending on the result of the comparison.

  Advantageously, the invention does not require having to define several thresholds related to the memory space of the first communication device to manage the problem of exceeding the memory capacity of this first device.

  Thanks to the invention, the memory space of the first device 15 is thus managed in a manner that is transparent to the user of this apparatus, insofar as the quality of service (for example, the image rate per second and the quality of the the video in the case of video data) remains unchanged.

  Moreover, such management causes a relatively short calculation time and low bandwidth consumption.

  According to one characteristic, when the size of the last data packet received is greater than the size of the memory space available in the first device, then the method comprises a step of transmitting, in the direction of the second device, a message of temporary interruption of the transmission of data packets to the first device.

  According to another feature, the interrupt message includes information indicating to the second apparatus the next data packet to be transmitted to the first apparatus.

  This allows, when the size of the last received packet is larger than the size of the available memory space in the first device, not to lose a packet, since the interrupt message will then indicate that the last received data packet by the device and that could not be stored will be forwarded to this device soon.

  Note that the interrupt message may alternatively indicate the last packet received.

  According to one feature, the method comprises the following steps: comparing the size of the unread memory space by the first device for storing video data packets with a threshold TH defined as follows: TH = min (RF x FR x N, Kx BS), where RF denotes the ratio of the maximum size of the memory space to the total number of images constituted by all the packets of video data to be transmitted from the second device to the first device, FR is the number of frames per second, N and K are constants and BS is the maximum size of the memory space, - depending on the result of the comparison, decision on the transmission to the second device, a message for resuming the transmission of data packets to the first apparatus.

  Note that here the invention provides for defining a threshold, unlike the prior art which provides for two, in order to know when it is necessary to send a message to resume the transmission of packets. data.

  According to one characteristic, when the size of the memory space not read in the first device is below the threshold TH, then the method comprises a step of transmitting, in the direction of the second device, a transmission resume message. data packets (RESTART) to the first device.

  In another aspect, the invention also provides a method of managing multimedia data packet transmissions between a data source and a plurality of communication devices connected to that source via a communication network, characterized in that the method includes a step, performed by the source, of receiving from an apparatus an interrupt message (PAUSE) for temporarily interrupting the transmission of data packets from the source to that apparatus , this message signifying that the size of the last data packet received by this device was greater than the size of the available memory space in this device for the storage of packets.

  Correlatively, the invention relates to a device for managing transmissions of multimedia data packets between a data source and several communication devices connected to this source via a communication network, characterized in that the source comprises receiving means from an apparatus of an interrupt message (PAUSE) for temporarily interrupting the transmission of data packets from the source to that apparatus, which message means that the size of the last packet data received by this device was greater than the size of the available memory space in this device for packet storage.

  Thus, following the analysis carried out in one or more of the communication devices connected to the data source, the latter provides for interrupting the transmission of data packets to one or more of these devices when it receives them. an interrupt message.

  According to one characteristic, the method comprises a step of deciding on the transmission of a data packet to the apparatus from which the interruption message originates and which notably takes into account the cumulative delay (OFFSET) during which the apparatus received no data packets from the source as a result of the transmission by this apparatus of one or more interrupt messages.

  The knowledge by the data source of this cumulative delay for a communication device is very useful insofar as this delay reflects the occupancy rate of the memory space of the device reserved for the storage of the data packets.

  Thus, if the accumulated time for a device is high, it means that the memory space is full or often full and the client machine has transmitted one or more interrupt messages to the source.

  If the cumulative time of a device is relatively small, it means that the client machine has relatively large memory space compared to the network bandwidth and / or the total size of the video.

  However, this may also mean that the rate at which packets are transmitted is small compared to bandwidth and memory space. Finally, it can also simply mean that the capabilities of the client machine are high, that is, the received packets are processed almost instantaneously.

  According to another characteristic, the method comprises a prior step of assigning the apparatus concerned a priority as a function, in particular, of the cumulative delay of this apparatus, the decision step taking into account this priority.

  Thus, the priority assigned to a device reflects the occupancy rate of the memory space of this device reserved for storing data packets.

  According to one characteristic, the method comprises a step of determining, at the current instant (ta), a parameter called margin (ta), representative of a quality of services offered to the apparatus, and which is written: margin (ta) = to + OFFSET + tp - ta, where o corresponds to the moment at which the device connected to the data source and tp corresponds to the instant at which at least one data packet is to be viewed by the device.

  The difference tp-ta is representative of the state of the device: if tp-ta is positive, it means that the device is in advance (the device has a sufficient number of data to process and it is affected therefore a low priority), while if tp-ta is negative, it means that the device is late (the device has insufficient data to process, which may penalize its operation and it is therefore assigns a high priority).

  When a PAUSE interrupt message is output from the unit, the range of this unit decreases. One is thus tempted to believe that the apparatus becomes less and less in advance, whereas this decrease is in fact due to the pause time, which falsifies the estimate of the state of the apparatus.

  To remedy this, the previously defined OFFSET parameter intervenes to shift the reference time to so that tp-ta continues to reflect the state of advance or delay of the apparatus.

  According to another characteristic, the method comprises the following steps: - determination of a parameter called variation which is written: variation = margin (ta) - margin (tb), o margin (ta) and margin (tb) are the parameters representative of the quality of services offered to the device at two different times ta and tb, with ta> tb, - assignment to the device of a priority according to the previously determined variation parameter.

  The parameter called variation thus makes it possible to estimate the evolution between the estimation of the current quality of service (instant ta) and the estimate of a previous quality of service (instant tb).

  Depending on the evolution of the estimation of the quality of service proposed, one chooses or not to change the priority assigned to a communication device.

  According to one characteristic, an interrupt message includes information indicating to the source the next packet of data to be transmitted to the device that transmitted this message.

  According to another characteristic, the source comprising a memory for storing the packets to be transmitted and a pointer for the last transmitted packet, when the source receives from a device a message (RESTART) for resuming the transmission of data packets to be transmitted. the source to this device, the pointer is repositioned on the data packet that is to be transmitted to the device.

  According to yet another aspect, the invention is directed to a method for managing transmissions of multimedia data packets between a data source and a plurality of communication devices connected to this source via a communication network, characterized in that the method comprises a step, performed by the source, for transmitting data packets to one or more apparatus, taking into account, in particular for each apparatus, the accumulated delay (OFFSET) during which the apparatus has not received any packets of data from the source following the transmission by this device of one or more interrupt messages, such messages intended to temporarily interrupt the transmission of data packets by the source.

  Correlatively, the invention is directed to a device for managing transmissions of multimedia data packets between a data source and a plurality of communication devices connected to this source via a communication network, characterized in that the source comprises means for transmitting data packets to one or more apparatus, taking into account, in particular for each apparatus, the cumulative delay (OFFSET) during which the apparatus has not received data packets from the source as a result of the transmission by this apparatus of one or more interrupt messages, such messages to temporarily interrupt the transmission of data packets by the source.

  Thus, as mentioned above, knowledge of the cumulative delay of one or more devices connected to the data source makes it possible to manage the resumption of transmissions to this or these devices.

  The invention also relates to a communication apparatus comprising a device as briefly described above.

  According to another aspect, the invention also aims at: computer-readable information storage means or a microprocessor comprising code instructions of a computer program for carrying out the steps of the method according to the invention as briefly discussed above, and - a partially or fully removable computer-readable information storage medium or microprocessor including code instructions of a computer program for executing steps of the process according to the invention such as that briefly described above.

  According to yet another aspect, the invention is directed to a computer program loadable in a programmable apparatus, comprising sequences of instructions or portions of software code for carrying out the steps of the method according to the invention as briefly described herein. above, when said computer program is loaded and executed on the programmable apparatus.

  The characteristics and advantages relating to the device, to the communication device comprising such a device, to the information storage means and to the computer program being the same as those described above concerning the method according to the invention, they will not be recalled here.

  Other characteristics and advantages of the present invention will emerge more clearly on reading the description which follows, made with reference to the appended drawings, in which: FIG. 1 schematically represents a client-type communication architecture; server in which the invention is implemented; FIGS. 2a, 2b and 2c schematically illustrate the operation of a shared memory in the server of FIG. 1; FIGS. 3a, 3b, 3c and 3d schematically illustrate the operation of a shared memory in the client machine of FIG. 1; FIG. 3e represents a transmission management algorithm according to the invention which is implemented by a client machine; FIG. 4 represents an algorithm for processing requests from communication devices such as that of FIG. 1 (client machine) and requesting the transmission of video data; FIG. 5 represents an algorithm for managing the transmission of data to different communication devices according to the priorities assigned to these devices; FIG. 6 represents an algorithm for dynamic assignment of priorities to communication devices according to the invention and which corresponds to step E503 of the algorithm of FIG. 5; - Figure 7 is an embodiment of a programmable apparatus embodying the invention.

  FIG. 1 shows a communication architecture 25 of the client-server type in which the invention is advantageously implemented.

  In this figure, a first communication device constituting a multimedia digital data source 1 called server is connected to several communication devices, of which only one referenced 2 and called 30 client machine is shown.

  Source 1 is connected to this apparatus via a communication network 3 and connections which are considered to be established for this machine.

  As will be seen in the description of FIGS. 5 and 6, the invention is not limited to a configuration where only one communication apparatus is connected to the data source.

  In the context of the invention, the communication apparatus requires from source 1 (server) digital multimedia data.

  The network is for example a wired 1 OObT Ethernet network. The network may also be, for example, a wireless local area communication network within a dwelling.

  More particularly, this wireless network complies with the IEEE802.11lb standard in the case of a so-called Ad-Hoc network, that is to say without centralization of data exchanges.

  In this example, this is a point-to-point data transmission between the server 1 and the client machines 2, 3 and 4.

  In order to implement these multiple transmissions, it is necessary to take into account, on the one hand, the management of the adaptation and transmission processes in parallel with the different client machines of the different adapted versions of the data requested and, on the other hand, , multiple accesses to the server memory (s).

  It should be noted that the data stored in the server may have been received from the environment outside the dwelling, for example, by another communication network such as the lnternet.

  The data grouped under the term multimedia data may be, in a nonlimiting manner, still images, videos, sound, text data (eg graphic documents ...), documents in HTML, signals from a fax machine or a printer ... In the exemplary embodiment considered, only 30 of the

  video data and these will be transmitted without loss on the network 3 in the form of data packets that are transmitted one by one.

  It will also be assumed that the transmission time of the data on the network is negligible.

  It should be noted that the invention also applies when the transmission time on the network is determined or determinable.

  The server 1 (data source) is a transmission management device in the sense of the present invention according to a first aspect.

  Note however that the server and the management device according to the invention may be separate elements.

  In this case, the different units of the server of FIG. 1 which serve to implement the invention then form part of the device according to the invention.

  The server 1 comprises a unit 11 temporarily storing video data to be transmitted to the communication apparatus 2.

  Useful information to be transmitted for reconstructing a video at a communication apparatus is stored in packets in the unit 11. Each packet thus contains the data enabling the destination apparatus to uncompress one or more images. of the video.

  Each of the packets also contains a number and a value tp which represents the moment at which the image or images contained in this packet are to be viewed by the recipient apparatus. The value of this instant will be, for example, equal to the ratio between the number of the first frame of the packet and the number of frames per second FR.

  It should be noted that the videos are created, for example, by a digital video camera or any other means of data acquisition or are transmitted by an external communication network and are stored in the unit 11.

  The schematic operation of this unit is shown in Figures 2a, 2b and 2c which will be described later.

  In addition, the videos are stored either in uncompressed form, for example in a YUV format, or in compressed form.

  It should be noted that a YUV video format means that the video concerned has three components, the Y component for the luminance and the U and V components for the chrominance.

  The server 1 comprises a scheduling unit 12 (known in English terminology as the "scheduling unit") of the data packets which aims to manage the transmission of packets between the device 1 and the client machines.

  The management of transmissions passes, for example, by a dynamic assignment of priorities to the different client machines in terms of whether or not a transmission of data packets.

  This assignment is dynamic insofar as it is established for each data packet that the server 1 has prepared for possible transmission to one of the client machines.

  For the next packet destined for the same client machine, the priority 15 assigned may indeed be different according to the circumstances. It should be noted that the dynamic character of the scheduling of the data packets has the advantage that new client machines can connect to the device 1 (server) while the scheduling of the packets intended for the client machines already connected is in progress.

  The various operations performed by the unit 12 will be detailed later in the description made with reference to FIGS. 5 and 6.

  The server 1 also comprises a unit 13 for receiving messages from the one or more client machines connected to the server, this unit also receiving requests from the latter.

  The server 1 also comprises a unit 14 which makes it possible to transmit data packets to the client machines through the network 3.

  The server 1 according to the invention further comprises a control unit 15 of the various operations performed.

  It is considered that the different client machines connected to the server 1 all have the same structure and operation as those

  of the machine 2 whose description will follow.

  The client machine 2 comprises a digital data reception unit 20.

  This unit 20 makes it possible to perform the reception of data coming from the server 1 and their transmission to the storage unit 23.

  The schematic operation of the storage unit 23 is shown in Figures 3a, 3b and 3c which will be described later.

  The client machine 2 also comprises a unit 21 for transmitting messages and requests through the network 3, and in particular to the server 1.

  This unit is mainly used to transmit to the server 1, on the one hand, messages called PAUSE intended to temporarily interrupt the transmission of data packets from the server to the client machine and, on the other hand, messages called RESTART which aim to resume the process of packet transmission by the server.

  The client machine 2 further comprises a unit 22 for processing the data packets stored in the unit 23 and a control unit 24 similar to the unit 15 of the server.

  The processing unit 22 retrieves the data contained in the packets stored in the unit 23 as they are received by the latter and are decompressed before being viewed.

  It should be noted that if the storage unit 23 contains no data, the last image displayed is frozen, which makes it impossible to view all the video requested by the client machine.

  If, on the other hand, the unit 23 contains data, two cases may occur: either the quantity of data present in the unit is sufficient to satisfy the constraints of the processing unit, or this quantity is insufficient.

  The constraints of the processing unit 22 are expressed, for example, in the number of frames per second because, indeed, a given video format specifies at what speed the video must be executed (for example, at 30 frames per second). second).

  Thus, in the case of a video at 30 frames per second, if each packet received by the storage unit 23 contains the constituent data of an image of the video, then the unit 23 must contain at least 30 packets for that the client machine 2 is considered to be one second ahead.

  If the unit 23 contains only 15 packets, the client machine is considered to be in advance of 0.5 seconds and so on.

  The constraints of the processing unit of a client machine correspond, for this machine, to a quality of services that the client expresses when it requires data from the server 1. The client expects 10 in return for receive data packets whose quality of service associated with each packet is consistent with the requested quality or, at least, as close as possible to it.

  Other constraints than the number of images per second can be expressed through the requested quality of services.

  Thus, there may be mentioned, for example, the time after which the client machine must receive a data packet with respect to the instant of issue of the request from this machine.

  It should be noted that, in a second aspect, the client machine 2 is also a transmission management device within the meaning of the present invention.

  FIGS. 2a, 2b and 2c, the description of which follows, illustrate the operation of the storage unit 11 of the server 1 of FIG. 1.

  Transcoding performed in the server generates data packets that are written in the shared memory of Figs. 2a and 2b, as indicated by the arrow shown on the left of Fig. 2b.

  Indeed, in this representation, data packets are stored in the memory 11 from left to right.

  The memory functions as a stack and the flag Pa acts as a pointer which represents the number of the next data packet to be transmitted to the client machine 2 of FIG.

  As shown in FIG. 2a, the memory 11 comprises several zones 30 to 34 which respectively contain data packets Pi, P2, P3, P4, P5, and a zone 35 for storing new packets.

  FIG. 2a shows the writing of a new packet P6 in a memory zone 36, thus leaving a memory area 37 empty to store new packets.

  This memory is accessed in read mode, that is to say by moving from the right to the left, when the scheduling algorithm of FIG. 5, which will be described later, is implemented.

  As will be seen later, depending on the priority assigned to a remote communication apparatus, the scheduling algorithm provides during a step E505 to send the data packet Pa to the considered client machine and to do this, read the memory 11 of Figure 2c as indicated by the arrow "sending".

  Once the packet has been transmitted to the client machine under consideration, the flag Pa is incremented by one unit, as shown in the comparison of FIGS. 2b and 2c.

  Note that a certain number of data packets that have been sent, namely the packets P1, P2, P3, are always stored in respective memory areas 30, 31, 32 of the memory 11 in order to be retransmitted in if necessary, for example, if one of these data packets was not received by the client machine for which it was intended.

  The server 1 will be informed of the packet to retransmit later by receiving the interrupt message from the client machine considered.

  It can, for example, be considered that after sending the next data packet P4, the previous packets will be successively deleted as new packets are written in the memory zone 37.

  Note that any data packet 30 stored in the memory 11 can be sent.

  The size of this memory is, for example, equal to the size of a compressed video sequence which is, for example, of the order of 2 to 3 Mbits.

  This memory is, for example, a buffer memory which can be of circular type.

  FIGS. 3a to 3d illustrate the operation of a part of the storage unit 23 of the client machine 2 of FIG. 1 which will be described in parallel with FIG. 3e.

  The algorithm of FIG. 3e comprises different instructions or portions of software code corresponding to steps of the transmission management method according to the invention implemented by the client machine.

  The computer program noted "ProgO" which is based on this algorithm is stored in the storage unit 23 and executed by the unit 22 under the control of the control unit 24.

  When the client machine 2 takes the form of the apparatus of FIG. 7, then this program is stored therein, for example, in the read-only memory of this apparatus.

  The portion of the storage unit 23 is, for example, a shared circular memory to which the data receiving unit 20 reads from the left as shown in Fig. 3a.

  In this figure, there is shown a filled memory area marked 40 on the right of the figure and a memory area 42 left empty to store 20 new packets received (available memory space).

  It will be noted, unlike the structure of the memory shown in FIGS. 2a to 2c, that here the memory does not maintain the structure of the packets in specific memory areas.

  However, like the memory of FIGS. 2a to 2c, the memory 23 also functions as a stack.

  When new packets are received by the receiving unit 20 of the client machine 2, the received packet data is written to the memory 23, as indicated by the left arrow in Fig. 3a.

  In addition, the corresponding packet number is stored to the extent that the size of this packet is less than or equal to the available memory space 42.

  FIG. 3b shows the memory that has just stored the last data received.

  In this state, the occupied memory space 44 occupies almost all of the total memory space and the available memory space 46 is greatly reduced.

  In such a case, when the next data packet is received (step E300) by the client machine 2 and its size (step E301) is considered to be greater (comparison step E303) than the size of the space available memory 46 (step E302), then the memory is considered as full (high rate of memory occupancy).

  In this case, the data of the received packet (decision step E304) is not written and the client machine 2 transmits the interrupt message PAUSE (step E305) to the server 1 via the transmission unit 21.

  This message contains the number of the last packet that was written in memory, which allows the server to know which is the next packet to retransmit to the client machine concerned.

  On the contrary, when the size of the received packet is greater than that of the available memory space, then it is decided (E304) to store the packet (E306), before reading it (E307) and then processing it (E308) .

  This memory functions as a FIFO type memory, that is to say "first in, first out" (known in English terminology under the term "First In First Out").

  The processing unit 22 of the client machine accesses this memory 25 in reading mode on the right in the figures by reading a number B bytes.

  When a number B of bytes has been read, the indicator or pointer d is incremented by a value B (d = d + B), as represented in FIG. 3c.

  Note that when the processing unit which wants to decode and display the video data needs to read B bytes and the memory does not contain a sufficient number of unread bytes, then this processing unit waits until a number B of unread bytes is available in memory.

  However, this is an extreme case that should not be encountered in practice because it penalizes, indeed, the rate at which the video sequence is to be viewed.

  When an interrupt message has been previously sent by the client machine following the observation of the high occupancy rate of the memory shown in FIG. 3b, then no packet newly received by the machine is written in memory.

  Since the processing unit 22 continues to read data therein (E307), the occupied memory space 44 of FIG. 3b is reduced to the memory space 48 of FIG. 3c and the memory space available, on the contrary, expands to move from the memory space 46 of Figure 3b to the memory space 50 of Figure 3c.

  As shown in FIG. 3d, a threshold TH is defined (step E309) as follows: TH = min (RF x FR x N, K × BS), where RF denotes the ratio of the maximum size of the memory on the total number of images constituted by all the video data packets constituting the video sequence in question, FR is the number of frames per second, N and K are constants and BS denotes the maximum size of the memory.

  It should also be noted that FR and the total number of images of the video sequence have previously been sent to the client machine when sending the first data packet of the video sequence.

  Thus, when the interrupt message PAUSE has been previously sent, then the size of the occupied memory space 48 in the memory 23 is compared with this threshold and, more particularly, the number of the memory space (step E311) is compared. Unread bytes in memory space 48 (E310) with respect to this threshold.

  When the number of unwanted bytes is less than the threshold TH, then it is decided (E312) to transmit (E313) a message of resumption of the transmission of RESTART data packets to the server 1 via the communication unit. transmission 21 of the client machine 2.

  The algorithm then starts waiting to receive a new packet (E300).

  It will be noted that the comparison with the threshold TH (step E31 1) is performed with respect to the number of unread data in the memory space.

  Data that has been read can always be stored in this memory space or deleted.

  On the contrary, when the number of unread bytes is greater than the threshold TH, then it is not necessary to transmit to the server a message of resumption of the transmission of packets and the algorithm goes on standby (E314) the time that the data is read.

  It is also conceivable that the number of unread bytes in the memory space 48 becomes smaller than the threshold TH without sending the transmission resume message.

  However, in certain circumstances, it is possible to block the visualization of the video sequence, because of the lack of data to feed the decoder.

  Note also that the buffer memory 23 shown in Figs. 3a to 3d constitutes a network memory which serves to store the data packets received from the network. There is another buffer specific to the decoder and which is used for the decoded data waiting to be viewed.

  FIG. 4 illustrates an algorithm comprising different instructions or portions of software code corresponding to steps of the method of processing requests from client machines, such as machine 2 of FIG. 1, and generating data packets to be transmitted.

  This algorithm illustrates the behavior of the server as soon as a client machine connects and a video sequence is requested.

  The computer program noted "Prog1" which is based on this algorithm is stored in the temporary data storage unit 11 of Figure 1 and executed by the unit 12 under the control of the control unit 15, which allows thus to implement the method according to the invention.

  This program is also stored in the apparatus of Figure 7.

  The algorithm of FIG. 4 comprises a first step denoted E400 5 during which the server 1 of FIG. 1 and, more particularly, the reception unit 13 receives a request sent by the client machine 2 from its unit. transmission 21.

  For example, this request is in the form of a character string for uniquely identifying video data stored in the storage unit 11 of the server 1 and necessary for the client machine 2.

  The request further specifies the quality of services requested by the client machine 2 and which must accompany the required data.

  It is considered that a data packet constituting part of the requested video sequence is present in the storage unit 11 as soon as it has to be transmitted.

  In the next step E401, several data counters are initialized, including a data packet counter Pa which is initialized to a value of 1.

  Also set is a priority value counter Pr which will subsequently be assigned to the different client machines (Pr = PRIOMAX where PRIOMAX is the highest priority value), a PAUSE indicator (PAUSE = false) and an OFFSET variable (OFFSET = 0) .

  It should be noted that the priority assigned to the various client machines 25 varies between a PRIOMIN low priority and a PRIOMAX high priority which are, for example, respectively equal to 1 and 30.

  The value of the priority that is assigned to each client machine translates, for the latter, the need to receive from the device 1 (server) data packets.

  As will be seen later, the decision to assign a priority value to a client machine depends on a dynamic analysis of the quality of services proposed for each data packet to be transmitted to this machine.

  The PAUSE indicator is a binary value which can take the values "true" or "false" and which changes according to the messages received by the reception unit 13 of the server.

  At initialization, the PAUSE variable takes the value "false" and takes the value "true" as soon as a message PAUSE is received by the server 1.

  It should be noted that each PAUSE interrupt message includes information informing the server of the next packet of data to be transmitted to the client machine at the origin of the message.

  For example, the message PAUSE contains the number of the last data packet written in the storage unit 23 of the client machine 2.

  Thus, upon receipt of a PAUSE message, the flag Pa is repositioned to the value of the last packet of data written in the unit 23.

  In this way, the server knows that the next packet to be transmitted to the client machine will be the packet whose flag is Pa + 1, and the pointer will be repositioned when receiving a transmission resume message.

  Moreover, upon receipt of a PAUSE message, a timer is initialized and the timer time value of the server Ts is stored.

  When a RESTART message for resuming transmission of data packets from the server to the client machine is received by the server, the corresponding instant Te of reception of this message is stored and the delay OFFSET is then calculated.

  This variable, which will be used in Figure 6, counts the cumulative delay during which the client machine is in PAUSE mode and therefore does not receive packets.

  The OFFSET delay is written as: OFFSET = OFFSET + Te-Ts. In step E401, OFFSET = 0.

  In addition, as soon as the RESTART message is received, the PAUSE variable is set to "false" and held until the next time a PAUSE message is received.

  The step E401 is then followed by a step E402 during which the list L of the client machines connected to the server 1 of FIG. 1 is updated by adding the new client whose request has just been received (step E400 ) and for which a first data packet has been stored in the storage unit 11 of Figure 1.

  The above-mentioned list L contains the enumeration of all the machines 10 (clients) that are connected to the server 1 as well as the priority assigned to each one of them.

  As soon as a client machine o, in other words, a client is added to this list, it is assigned the highest priority (PRIOMAX).

  It should be noted that for each client machine in the list L are associated the four aforementioned variables: Pa, Pr, OFFSET and PAUSE.

  The use of this list will also be detailed later in the description made with reference to FIGS. 5 and 6.

  The next step E403 indicates the receipt by the server of a PAUSE or RESTART message transmitted by one or more connected client machines.

  During the next step E404, the variable PAUSE changes its value according to the message received, the value Ts or Te is stored, the indicator Pa changes or the delay OFFSET is calculated as appropriate.

  It should be noted that steps E403 and E404 may occur at any time during connection of the server to the client machines.

  In a subsequent step E405, the server transmits the packet or packets stored in the storage unit 11 according to the scheduling algorithm shown in FIG. 5.

  When the or all packets have been transmitted to the client machine 2, the transmission is terminated and the client machine 2 is disconnected from the server 1 (step E406).

  The next step E407 provides an update of the list of client or client machines during which the client machine that has just been disconnected in step E406 is deleted, which terminates the algorithm of FIG. 4.

  It will be noted that the algorithm of FIG. 4 is executed for each new client machine connected to the server 1.

  Since several client machines can be connected at the same time, the algorithm of FIG. 4 is executed in parallel to generate data packets for a given client machine.

  It should also be noted that when a client machine is connected to the server 1, it is necessary to always have at least one data packet to be transmitted, when such a transmission is decided by the algorithm of FIG. we will see later.

  FIG. 5 illustrates an algorithm comprising different instructions or portions of software code corresponding to steps of the method according to the invention.

  The computer program noted "Prog2" which is based on this algorithm is stored in the temporary data storage unit 11 of FIG. 1 and executed by the scheduling unit 12 of this figure, under the control of the control unit 15, which thus allows the method according to the invention to be implemented.

  This program is also stored in the apparatus of Figure 7.

  When executing the algorithm of FIG. 5, the server 1 of FIG. 1 manages the data transmissions between this device and the different client machines connected to it, according to the priorities that are assigned to these machines. .

  It should be noted that the scheduling process of the data packets whose algorithm is shown in FIG. 5 is implemented as soon as the server 1 of FIG. 1 operates.

  The process whose algorithm is shown in Figure 4, meanwhile, is implemented on receipt of a request from a client machine.

  Thus, these two processes are executed in parallel and the process of which the algorithm is represented in FIG. 5 uses that of FIG. 4 to recover a data packet generated by the latter and intended for a client machine, when is decided to make such a transmission to said client machine.

  The algorithm of FIG. 5 starts with an initialization step E500 during which the PRIOMAX maximum priority value is assigned to a loop index noted p. During the next step noted E501, a variable c of the list L of the client or client machines referred to above is initialized to the value 1, which corresponds to the first element of the list L. The following step E502 provides for performing a test to test the value of the PAUSE indicator of the current client machine c and check whether it is equal to the value "true".

  If so, this means that a PAUSE interrupt message has been received from this machine and step E502 is followed by step E507. No packet will be transmitted to the machine in question.

  If not, the algorithm continues with a step E503 in which the priority Pc associated with the current client machine is updated.

  The description of this step E503 will be discussed in more detail later in the description made with reference to FIG.

  In this step, the current client machine is dynamically assigned a priority adapted to the results of the analysis performed by the management device according to the invention.

  The next step E504 consists of a test in which it is determined whether the priority assigned to the current client machine, Pc, is greater than or equal to the value of the aforementioned p-loop index.

  If so, step E504 is followed by a step E505 in which unit 12 of FIG. 1 retrieves the data packet from the storage space 11 of FIG. Pa which was generated during the execution of the algorithm of FIG. 4 in order to be transmitted to the current client machine, for example the machine 2 of FIG. 1.

  This packet is then transferred to the transmission unit 14 of FIG. 1 and is then transmitted via the communication network 3 to the client machine in question.

  The indicator Pa of the memory 11 is then incremented by one unit 10 (FIGS. 2b and 2c).

  Step E505 is then followed by step E506.

  Thus, a high priority has been assigned to the current client machine so that no data packet is transmitted to the latter, following the analysis of the connection between the device 1 and the current client machine 15 considered and the step E504 is then followed by step E506.

  During the next analysis of this same connection, the result of this analysis can change and the current client machine can then be assigned a priority such that a data packet will be transmitted to it.

  In step E506, a test is performed on the value of the variable c with respect to the list L (c = C max, where C max represents the last element of the list L) to determine whether all the machines client connected to the server 1 were processed during the execution of the loop of the algorithm of FIG.

  If not, step E506 is followed by step E507 in which variable c is incremented by one so that the loop of the algorithm of FIG. 5 consists of steps E503, E504, E505, E506 is again executed for the next client machine of the list L. On the contrary, when the result of the test practiced in step E506 is positive, then the latter is followed by step E508.

  During this step, another test is performed to determine if the loop index p has reached the minimum value PRIOMIN.

  If not, the step E508 is followed by a step E509 in which the p-loop index is decremented by one unit and then all the client machines connected to the server are processed again from the first one (c = 1) and executing the loop formed by steps E502 to E506.

  If the test performed in the step E508 leads to a positive result, the latter is then followed by the previously described step E509 during which the loop index is reset to the PRIMPAX high priority value and the The algorithm of FIG. 5 is again executed.

  It should be noted that the execution of the process whose algorithm is represented in FIG. 5 depends on the content of the list L of client machines (or clients) connected to the device 1 of FIG. 1: as soon as a client machine is connected the process is executed and it stops when no client machine is connected.

  We will now describe the dynamic priority assignment process for a current client machine marked by the variable c in the list L defined above, the connection of this client machine with the device 1 of FIG. execution of the algorithm of FIG.

  The algorithm of FIG. 6 is part of the algorithm of FIG. 5 and comprises for this purpose different instructions or portions of software code corresponding to steps of the transmission management method according to the invention.

  The algorithm of FIG. 6 starts with a step E600 which plans to estimate the quality of services offered to the current client machine c for a data packet that the device 1 of FIG. 1 proposes to transmit to it.

  Note that for the purposes of updating the priority Pc assigned to step E502 of Figure 5, it is necessary to know the contents of the packet to be transmitted.

  Note also that this estimation step takes into account the time tp after which the packet must reach the current client machine.

  This information is known from the device 1 according to the invention (server) through the quality of services expressed in the request from this client machine.

  It will also be noted that the time tp is somewhat representative of the filling rate of the storage unit of the client machine under consideration.

  This time depends on the quality of services expressed in the request from the client machine considered, as well as the content of the data packet to be transmitted to this machine.

  Indeed, when, for example, the requested quality of services is expressed in the number of images per second, according to whether the images are coded in Intra or Inter mode, for the same packet size, the number of images contained in a packet is higher in the case of an image coded in Inter mode.

  Thus, knowing the quality of services expressed by the client machine (eg number of frames per second) and taking into account the content of the packet (number of images per packet), the device 1 according to the invention is able to determine the time above mentioned.

  More particularly, during step E600, a parameter called margin (ta), which is representative of the quality of services offered to the current client machine considered for the data packet, is determined at a time ta, called the current time. concerned which will eventually be transmitted.

  This parameter is expressed as: margin (ta) = to + OFFSET + tp - ta, where o corresponds to the instant when the current client machine has connected to device 1.

  Once this estimation step has been performed, the next step E601 provides the determination of the evolution, over time, of the estimation of the quality of services offered to the current client machine.

* This determination is made for data packets to be transmitted at different times.

  For example, one can be interested in the transmission of two successive data packets over time, or even determine this evolution for data packets to be transmitted spaced apart from several packets in time.

  More particularly, during step E601, a parameter called "variation" is determined which is written as follows: variation = margin (ta) - margin (tb), where the parameters margin (ta) and margin ( tb) are representative of the service quality estimated at two different times ta and tb with ta> tb and for two different data packets.

  The variation thus makes it possible to estimate the evolution between the estimate of the quality of current services and the estimate of a previous quality of services.

  Step E601 is then followed by step E602 in which a test is performed to determine if the previously determined variation is zero.

  If so, the quality of services proposed has therefore not changed since the last estimate, it is chosen not to change the priority assigned to the current client machine c whose connection with the device 1 is being analyzed, which ends the algorithm of Figure 6.

  On the contrary, when the result of the test performed in step E602 is negative, then the next step E603 provides for another test to determine whether the previously determined variation is greater than or equal to the margin determined in step E600.

  This test makes it possible to detect a large variation in the estimation of the quality of services proposed for a data packet to be transmitted to the current client machine.

  Thus, when the result of this test is positive, the management device according to the invention is informed that the margin determined at the current time ta is decreasing and that it is therefore necessary to anticipate so as not to degrade the quality of services offered to the client machine considered.

  The step E603 is then followed by a step E607 in which a high priority and, moreover, a maximum priority, PRIOMAX, is assigned to the current client machine whose connection with the device 1 of FIG. 1 is in progress. analysis.

  This means that it is urgent to provide a transmission of a data packet to this client machine.

  This then puts an end to the execution of the algorithm of FIG.

  On the contrary, when the result of the test performed in step E603 is negative, the next step E604 provides for an additional test.

  During this test, it is determined, on the one hand, if the margin previously determined in step E600 is greater than or equal to a predetermined threshold and, on the other hand, if there is at least one client machine for which a negative margin has been detected (Nb> 0).

  The predetermined threshold corresponds, for example, to one second and makes it possible to detect whether the current client machine will be considered to be sufficiently far in advance, that is to say that the filling rate of its storage unit is sufficient to not interrupt the process of decoding and viewing the video data requested by this machine.

  It should be noted that it would have been sufficient to detect whether the margin determined in step E600 was positive to ensure that the quality of services estimated by the device 1, for a data packet to be transmitted to this device. machine client, was consistent with the quality of services requested by the latter.

  However, this one verification alone is not enough to determine whether a client machine needs a data packet, since a positive margin at one instant may become negative at the next instant and thus translate a degradation. the quality of services offered to this machine.

  When the result of the test practiced in step E604 is positive, that is to say when the estimation of the quality of services offered to a current client machine for a data packet is in accordance with the quality of services requested by the latter and when, for at least one other client machine, the estimation of the quality of services proposed for a data packet to be transmitted to the latter does not comply with the quality of services requested, then this step is followed by a step E605.

  During this last step, it is planned to assign to the current client machine a minimum priority Pc which is equal to PRIOMIN -1.

  The assignment of this minimum priority indicates to the device 1 according to the invention that it is not necessary to transmit a data packet to the current client machine c.

  This information is used in the algorithm of FIG. 5 during the test performed in step E504 which leads, in this specific case, directly to the execution of step E506, without transmitting a data packet as provided by step E505.

  The priority thus assigned to the current client machine excludes any transmission of data packet to this machine. This is of course valid only for the estimate which has just been made of the connection established between the device 1 and the current client machine considered and is likely to be called into question during the new estimation, for a new data packet to transmit.

  On the contrary, when the result of the test practiced in step E604 is negative, then this step is followed by a step E606.

  This case occurs either when the margin determined in step E600 is not greater than the threshold set or when no negative margin has been detected for other client machines.

  In step E606 a test is made to determine if the margin of the current client machine whose connection to the device 1 according to the invention is being analyzed is positive.

  This amounts, in other words, to determine whether the quality of service offered to this client machine in relation to a data packet to be transmitted is consistent with the quality of services requested by this machine.

  When the result of the test practiced in step E606 is negative, which corresponds to a nonconformity of the quality of services proposed, as it was estimated, with the quality of services requested by the current client machine, then this step is followed by step E607.

  During this last step, it is planned to assign the current client machine the priority value Pc maximum (PRIOMAX).

  This informs the management device according to the invention that it is necessary to perform a transmission of a packet for this client machine when executing the algorithm of FIG. 5.

  It should be noted that steps E602, E603, E604 and E606 are steps in which the quality of services estimated at step E600 or its variation determined in step E601 are compared with the quality of services requested by the client machine. common.

  When the result of the test practiced in step E606 is positive, which means that the estimate of the quality of services offered to the current client machine is in accordance with the quality of services requested by the latter, then this step is followed of a step E608.

  During this last step a value called "step" is determined which corresponds to the step, also called priority jump, whose current priority assigned to the current client machine must be varied in order to take into account the estimation of the quality. previously performed. In this step E608, the increment of the priority to be applied to the current priority is determined as follows: step = | E (PRIOMAX (variation / margin)) l, o | x | denotes the mathematical function "absolute value of x" and E (x) denotes the mathematical function "integer part of x".

  In the next step E609 the value of the priority to be assigned to the current client machine is determined as follows: Pc = Pc - sign (not), where o sign (x) denotes the mathematical function "sign" of x.

  The execution of steps E608 and 609 thus makes it possible to adjust the value of the priority assigned to the current client machine according to the estimate of the quality of services performed previously in step E600.

  This makes it possible to refine the value of the priority assigned to the current client machine rather than systematically assigning an extreme value of the PRIOMAX or PRIOMIN type.

  The algorithm of FIG. 6 is terminated after execution of step E609.

  Note that the transmission of coded video data is "on the fly" (known in English terminology under the term "streaming"), that is to say that the information needed by the user to reconstruct an image are stored in the storage unit 11, then are transmitted to the client machine 2 before all the video data is coded.

  Thus, the user receives at the level of the client machine 2 the packet information, in each of which are information allowing this user to decode a current image.

  It should be noted that the invention also applies when the data source (server) which is connected to the communication apparatus does not receive requests from them and specifying a requested quality of service.

  Indeed, the server estimates, from the only information available to it, the quality of services it is likely to offer a device by transmitting one or more data packets.

  Based on this estimate, the server is then able to evaluate the future workload of the device, to do the same with other devices and assign priorities accordingly to the devices concerned.

  This can be done without the server knowing the quality of services requested by each device.

  In this case, the algorithms of FIGS. 4 to 6 must be modified accordingly.

  Referring to Figure 7, an example of a programmable communication apparatus embodying the invention is described.

  This apparatus comprises the transmission management device according to the invention shown in FIG. 1 and the operation of which is described with reference to FIGS. 2 to 6. According to one aspect of the invention, the programmable device is connected to several communication devices. communication playing the role of client machines, identical to the device 1 of Figure 1.

  In another aspect, the programmable apparatus is a client machine connected to the device 1 of the figurel.

  According to the embodiment chosen and shown in FIG. 7, an apparatus embodying the invention is for example a microcomputer 700 or a workstation connected to different peripherals, for example a digital camera 701 (or a scanner, or any other device). acquisition means or image storage) connected to a graphics card and providing the device with multimedia data to be processed and transmitted to one or more other devices according to the aspect of the invention in question.

  The apparatus 700 comprises a communication bus 702 to which are connected: a central processing unit 703 (microprocessor), which performs the function of the control unit 15 of FIG. 1, a read-only memory 704, which can comprise the programs "ProgO", "Prog 1" and "Prog2", - a RAM 706, having registers 707 adapted to record variables and parameters created and modified during the execution of the aforementioned programs, in particular, p, c, Pr, Pa, Cmax, PAUSE, PRIOMAX, PRIOMIN, margin, variation, threshold, Nb, not, TH mentioned with reference to the preceding figures, as well as the data packets generated during the execution of the algorithm of FIGS. and 5.

  a screen 708 making it possible to display data and / or to act as a graphical interface with the user who can interact with the programs according to the invention, using a keyboard 710 or any other means such as a pointing device, not shown, such as for example a mouse or an optical pencil, a hard disk 712 which may comprise the aforementioned "ProgO", "ProgI" and "Prog2" programs, a floppy disk drive 714 adapted to receive a floppy disk. 716 and to read or write data processed or to be processed according to the invention, - a communication interface 718 connected to a communication network 720, for example the Internet, the interface being able to transmit and receive Datas.

  In the case of audio data, the apparatus further comprises an input / output card connected to a microphone which are both unrepresented.

  The communication bus enables communication and interoperability between the various elements included in the microcomputer 700 or connected thereto. The representation of the bus is not limiting and, in particular, the central unit is able to communicate instructions to any element of the microcomputer 700 directly or through another element of the microcomputer 700.

  The executable code of each program allowing the programmable apparatus to implement the transmission management process of Figure 3e, the request processing and packet generation processes (Figure 4), dynamic assignment of priorities to Connected client machines (FIGS. 5 and 6) according to the invention can be stored for example in the hard disk 712 or in the read-only memory 704 as shown in FIG.

  According to one variant, the diskette 716 may contain data as well as the executable code of the aforementioned programs which, once read by the apparatus 700, will be stored in the hard disk 712.

  In the second variant, the executable code of the programs may be received via the communication network 720, via the interface 718, to be stored in the same manner as that described previously.

  Floppies can be replaced by any information medium such as, for example, a compact disc (CD-ROM) or a memory card. In general terms, computer-readable or microprocessor-readable information storage means, whether or not integrated into the device, possibly removable, is adapted to store one or more programs whose execution allows the implementation to be carried out. of the process according to the invention.

  More generally, the program or programs may be loaded into one of the storage means of the device 700 before being executed.

  The central unit 703 will control and direct the execution of the instructions or portions of software code of the program or programs according to the invention, which instructions are stored in the hard disk 712 or the read-only memory 704 or in the other elements of aforementioned storage. When powering on, the program or programs that are stored in a non-volatile memory, for example the hard disk 712 or the ROM 704, are transferred into the RAM 706 which will then contain the executable code of the one or more programs according to the invention, as well as registers for storing the variables and parameters necessary for the implementation of the invention.

  It should be noted that the communication apparatus comprising the device according to the invention may also be a programmed apparatus.

  This device then contains the code of the computer program or programs for example frozen in a specific application integrated circuit (ASIC).

Claims (38)

  1. A method of managing transmissions of multimedia data packets between at least two communication devices connected to one another, said first device, receiving data packets from at least one other device, said second device, characterized in that the method comprises the following steps performed by the first apparatus: comparing the size of the last data packet received from the second apparatus with the size of the available memory space in the first apparatus for storage, - depending on the result of the comparison, decision to transmit, in the direction of the second device, an interrupt message (PAUSE) to temporarily interrupt the transmission of data packets to the first device.   2. Method according to claim 1, characterized in that, when the size of the last received data packet is greater than the size of the available memory space in the first device, then the method comprises a transmission step, in the direction of the second apparatus, a message of temporary interruption of the transmission of data packets to the first device.   3. Method according to claim 2, characterized in that the interrupt message includes information indicating to the second device the next packet of data to be transmitted to the first device.   4. Method according to one of claims 1 to 3, characterized in that it comprises the following steps: - comparison of the size of the memory space unread by the first device for storing video data packets with a threshold TH defined as follows: TH = min (FRx FRxN, KxBS), where RF denotes the ratio of the maximum size of the memory space to the total number of images constituted by all the video data packets to transmit the second device to the first device, FR is the number of frames per second, N and K are constants and BS is the maximum size of the memory space, - depending on the result of the comparison, decision on the transmitting, in the direction of the second apparatus, a message for resuming the transmission of data packets to the first apparatus.   5. Method according to claim 4, characterized in that, when the size of the memory space not read in the first device is below the threshold TH, then the method comprises a step of transmission, towards the second device, of a message for resuming the transmission of data packets (RESTART) to the first apparatus.   6. A method of managing transmissions of multimedia data packets between a data source and several communication devices connected to this source via a communication network, characterized in that the source comprises receiving means from 15 of a device of an interrupt message (PAUSE) for temporarily interrupting the transmission of data packets from the source to this device, this message meaning that the size of the last packet of data received by this device was greater than the size of the available memory space in this device for packet storage.   7. Method according to claim 6, characterized in that it comprises a step of deciding the transmission of a data packet to the apparatus from which the interruption message comes and which takes into account in particular the cumulative delay (OFFSET). during which the apparatus did not receive data packets from the source as a result of the transmission by this apparatus of one or more interrupt messages.   8. A method according to claim 7, characterized in that it comprises a prior step of assigning to the apparatus concerned a priority (Pc) as a function in particular of the cumulative delay (OFFSET) of this apparatus, the step of decision taking into account this priority.   9. Method according to claim 7 or 8, characterized in that it comprises a step of determination at the current time (ta) of a parameter called margin (ta), representative of a quality of services proposed to the device, and which is written: margin (ta) = to + OFFSET + tp - ta, where o corresponds to the moment when the device connected to the data source and tp corresponds to the moment at which least one data packet must be viewed by the device.   10. Method according to claim 9, characterized in that it comprises the following steps: - determination of a parameter called variation which is written: variation = margin (ta) - margin (tb), o margin (ta) and margin (tb) are the parameters representative of the quality of services offered to the apparatus at two different times ta and tb, with ta> tb, assignment to the apparatus of a priority according to the previously determined variation parameter.   11. Method according to one of claims 6 to 10, characterized in that an interrupt message includes information indicating to the source the next data packet to be transmitted to the device that transmitted this message. 15 12. Method according to one of claims 6 to 11, characterized in that the source comprising a memory for storing the packets to be transmitted and a pointer to the last transmitted packet, when the source receives from a device a message (RESTART ) of resuming the transmission of data packets from the source to this apparatus, the pointer is repositioned on the data packet to be transmitted to the apparatus.   13. A method of managing transmissions of multimedia data packets between a data source and several communication devices connected to this source via a communication network, characterized in that the method comprises a step, performed by the source, of transmission of data packets to one or more devices, taking into account, in particular for each apparatus, the cumulative delay (OFFSET) during which the apparatus has not received any data packets from the source as a result of the transmission by this apparatus of one or more interrupt messages, such messages to temporarily interrupt the transmission of data packets 30 by the source.   14. Device for managing transmissions of multimedia data packets between at least two communication devices connected to one another, said first device receiving data packets from at least one other device, said second device, characterized in that the first apparatus comprises - means for comparing the size of the last data packet received from the second apparatus with the size of the memory space available in the first apparatus for storage, means for deciding on the transmission, in the direction of the second apparatus, an interrupt message (PAUSE) for temporarily interrupting the transmission of data packets to the first apparatus, depending on the result of the comparison.   15. Device according to claim 14, characterized in that the first apparatus comprises means for transmitting, in the direction of the second apparatus, a message temporarily interrupting the transmission of data packets to the first apparatus, when the size of the last received data packet is greater than the size of the available memory space in the first device Apparatus according to claim 15, characterized in that the interrupt message includes information indicating to the second apparatus the next data packet to be transmitted to the first apparatus.   17. Device according to one of claims 14 to 16, characterized in that the first apparatus comprises: means for comparing the size of the memory space not read by the first apparatus for storing video data packets with a threshold TH defined in the following way: TH = min (RFxFRxN, KxBS), where RF denotes the ratio of the maximum size of the memory space to the total number of images constituted by all the video data packets to transmit the second device to the first device, FR is the number of frames per second, N and K are constants and BS is the maximum size of the memory space, - means of decision as to the transmission, in the direction of second apparatus, a message for resuming the transmission of data packets to the first device, depending on the result of the comparison.   18. Device according to claim 17, characterized in that the first apparatus comprises means for transmitting, in the direction of the second apparatus, a message for resuming the transmission of data packets (RESTART) to the first apparatus, when the Unread memory space size in the first device is below TH threshold.   19. Device for managing transmissions of multimedia data packets between a data source and several communication devices connected to this source via a communication network, characterized in that the source comprises receiving means from 10 of a device of an interrupt message (PAUSE) for temporarily interrupting the transmission of data packets from the source to this device, this message meaning that the size of the last packet of data received by this device was greater than the size of the available memory space in this device for packet storage.   20. Device according to claim 19, characterized in that the source comprises means for deciding the transmission of a data packet to the apparatus from which the interrupt message is derived and which notably take into account the cumulative delay (OFFSET). during which the apparatus has not received any data packets from the source as a result of transmission by this apparatus of one or more interrupt messages.   21. Device according to claim 20, characterized in that the source comprises means for prior assignment to the concerned apparatus of a priority (Pc) depending in particular on the cumulative delay (OFFSET) of this apparatus, the decision means taking into account this priority.   22. Device according to claim 20 or 21, characterized in that the source comprises means for determining at the current time (ta) a parameter called margin (ta), representative of a quality of services offered to the user. device, and which is written: margin (ta) = to + OFFSET + tp - ta, where o corresponds to the moment when the device has connected to the data source and tp corresponds to the instant at which at least one data packet must be viewed by the device.   23. Device according to claim 22, characterized in that the source comprises: - means for determining a parameter called variation which is written: variation = margin (ta) - margin (tb), o margin (ta) and margin (tb) are the parameters representative of the quality of services offered to the device at two different times ta and tb, with ta> tb, - means for assigning the device a priority according to the parameter variation previously determined.   24. Apparatus according to one of claims 19 to 23, characterized in that an interrupt message includes information indicating to the source the next data packet to be transmitted to the apparatus which transmitted this message.   25. Device according to one of claims 19 to 24, characterized in that the source comprises: - storage means of data packets to be transmitted, a pointer to the last transmitted packet, - receiving means, from an apparatus, a message (RESTART) for resuming the transmission of data packets from the source to this apparatus, - means for repositioning the pointer to the data packet to be transmitted to the apparatus.   Device for managing the transmission of multimedia data packets between a data source and a plurality of communication devices connected to this source via a communication network, characterized in that the source comprises means for transmitting packets. to one or more devices, taking into account, in particular for each device, the cumulative delay (OFFSET) during which the device did not receive data packets from the source following the transmission by this apparatus of a or several interrupt messages, such messages to temporarily interrupt the transmission of data packets by the source.   27. Communication device, characterized in that it comprises a transmission management device according to one of claims 14 to 18.   28. Communication device, characterized in that it comprises a transmission management device according to one of claims 19 to 25.   29. Communication device, characterized in that it comprises a transmission management device according to claim 26.   A computer-readable or microprocessor-readable information storage means having code instructions of a computer program for carrying out the steps of the transmission management method according to one of claims 1 to 5.   31. A computer-readable or microprocessor-readable information storage means having code instructions of a computer program for performing the steps of the transmission management method according to one of claims 6 to 12.   32. A computer-readable or microprocessor-readable information storage means having code instructions of a computer program for performing the steps of the transmission management method according to claim 13.   33. Removable, partially or completely readable information storage medium by a computer or a microprocessor comprising code instructions of a computer program for carrying out the steps of the transmission management method according to one of the claims. 1 to 5.   34. Removable, partially or fully readable information storage medium by a computer or microprocessor including code instructions of a computer program for executing the steps of the transmission management method according to one of the Claims 6 to 12.   35. Removable, partially or fully computer readable information medium or microprocessor comprising code instructions of a computer program for performing the steps of the transmission management method according to claim 13.   36. Computer program loadable in a programmable apparatus, characterized in that it comprises sequences of instructions or portions of software code to implement the steps of the transmission management method according to one of claims 1 at 5, when this computer program is loaded and executed by the programmable device.   37. Computer program loadable in a programmable apparatus, characterized in that it comprises sequences of instructions or portions of software code to implement the steps of the transmission management method according to one of claims 6 to 12, when this computer program is loaded and executed by the programmable apparatus.   38. Computer program loadable in a programmable apparatus, characterized in that it comprises sequences of instructions or portions of software code to implement the steps of the transmission management method according to claim 13, when this program 10 computer is loaded and executed by the programmable device.